BUG: Fixed slicing of chararrays on Python 3.

When taking a slice of a chararray it was calling the rstrip() method
on the resulting slice, resulting in a new array rather than a view
of the original.  This was an unintended consequence of the sq_slice
member of the tp_as_sequence mapping being ignored in Python 3, so
that slice lookups go directly through __getitem__.

Fix test_expandtabs to not make the assumption that rstrip() will be applied twice to a value when using T[x][y] style indexing.

1 files changed, 3 insertions, 1 deletions

diff --git a/numpy/core/defchararray.py b/numpy/core/defchararray.py
index cc6cb5a38..92ea8209c 100644
--- a/numpy/core/defchararray.py
+++ b/numpy/core/defchararray.py
@@ -1849,12 +1849,14 @@ class chararray(ndarray):
 
     def __getitem__(self, obj):
         val = ndarray.__getitem__(self, obj)
-        if issubclass(val.dtype.type, character) and not _len(val) == 0:
+
+        if isinstance(val, character):
             temp = val.rstrip()
             if _len(temp) == 0:
                 val = ''
             else:
                 val = temp
+
         return val
 
     # IMPLEMENTATION NOTE: Most of the methods of this class are